履带式机器人的电子定向和双目视觉测距
|
摘要
履带式机器人具有特殊的机械结构，可以适应复杂的路面环境，它的应用也越来越广泛。本文利用陀螺仪实现机器人的电子定向，完成了机器人旋转角度的控制。陀螺仪是测量旋转物体角速度的一种内部传感器，可以测量出旋转物体本身姿态的变化，最为重要的是不需要外界的参考信号。利用陀螺仪的这个特性可以使机器人在能见度较低的环境下或者对环境不熟悉以及参照物不足的情况下实现比较精确的导航和定位。双目视觉测距中，采用了鱼眼镜头与标准镜头相配，且鱼眼镜头轴线垂直于移动机器人底面，而标准镜头轴线平行于移动机器人前进方向。从而利用鱼眼镜头可实现360°的环景观察，方便寻找目标物体，在电子陀螺仪支持下，利用标准镜头对目标物体进行识别和测距。这一独特优势能够使它在民用、军事、航空等那些依赖于视觉信息做出决策的各行业都具有非常重要的意义。
本文以履带式机器人为研究对象，在双目测距中，.完成了对鱼眼图像的中心标定。根据颜色特征对图像进行两次分割，得到灰度图像和二值图像，再将分割后的图像经过噪声消除和中值滤波的手段，自动快速的识别目标物体。在试验中，把物体近似看为一圆点，用形心来代替目标物体，并对双目测距的公式进行了推导。实验测量了多组数据，并进行了误差分析，建立了误差修正模型。
在电子定向中，设计完成了基于ENC-03陀螺仪的外围电路。让陀螺仪输出的模拟信号依次经过滤波、放大、A/D转换、变成数字信号，通过单片机和电脑相连。然后从计算机上读取陀螺仪的输出电压值，利用公式换算成角速度，并进行积分，从而求出物体旋转的角度。试验测量了多组角度数值，并对所求的角度进行验证，计算出最大误差，满足了ENC-03陀螺仪精度的要求。
通过无线网络，完成了计算机对履带式机器人的远程控制，实现了两者之间的通信。编写机器人软件控制界面，实时显示履带式机器人的偏转角度。
The tracked mobile robot with special mechanical structure, can adapt to a complex pavementenvironment, and its application is becoming more and more widely. We use the gyroscope toachieve electronic directional and the control of rotation angle based on the robot. Thegyroscope is an internal sensor which can measure the angular velocity of rotating objects. Itis very important that gyroscope does not need the external reference signal. We can usecharacteristic of the gyroscope to achieve more precise navigation and positioning when therobot is in a low environment or we are not familiar with the environment. In the distancemeasurement system of Binocular stereo vision，we use the fish-eye lens and the standard lens.The fish-eye lens axis is perpendicular to the bottom surface of the mobile robot and thestandard lens axis parallel to the direction of mobile robot. So we use the fish-eye lens to findthe target objet easier. At last, we use the standard lens to identify the target objet and achievethe distance measurement with the help of the electronic gyroscope.Because of the uniqueadvantage, it has a very important significance in civil, aerospace, military and others whichrely on the visual information to make decision.
This paper is based on tracked mobile robot. In the distance measurement system ofBinocular stereo vision, we complete the center calibration of the fish-eye image．Accordingto the characteristics of image color, we make the color segmentation twice and get grayscaleimage and binary image．Also we can identify the target object automatically and quicklythrough the image which has been eliminated noise and median filtering．In the experiment,we use centroid of the object to replace of the target object and deduce the formula ofbinocular stereo-vision ranging．At last, we make the analysis and correction of the erroraccording to the experimental data.
As the main part of this article, it designs the ENC-03 gyroscope peripheral circuit inelectronic directional. By using the filtering, amplifier, A/D conversion, it converts the outputanalog signals of the gyroscope into digital signals, and then links with single-chipmicrocomputer and computer. After that, the output voltage value of the gyroscope can beread from the computer. Then the value can be converted into angular velocity and integral, soas to find out the rotation angle of the object. Measuring the Angle of values, verifying theresult, and calculating the maximum error, it meets the accuracy requirement of the ENC-03gyroscope.
Through the wireless network, we can use the computer to control the tracked mobilerobot and complete the communication between the computer and the tracked mobile robot.We write the control software of the tracked mobile robot to display the deflection angle inreal time．
本文以履带式机器人为研究对象，在双目测距中，.完成了对鱼眼图像的中心标定。根据颜色特征对图像进行两次分割，得到灰度图像和二值图像，再将分割后的图像经过噪声消除和中值滤波的手段，自动快速的识别目标物体。在试验中，把物体近似看为一圆点，用形心来代替目标物体，并对双目测距的公式进行了推导。实验测量了多组数据，并进行了误差分析，建立了误差修正模型。
在电子定向中，设计完成了基于ENC-03陀螺仪的外围电路。让陀螺仪输出的模拟信号依次经过滤波、放大、A/D转换、变成数字信号，通过单片机和电脑相连。然后从计算机上读取陀螺仪的输出电压值，利用公式换算成角速度，并进行积分，从而求出物体旋转的角度。试验测量了多组角度数值，并对所求的角度进行验证，计算出最大误差，满足了ENC-03陀螺仪精度的要求。
通过无线网络，完成了计算机对履带式机器人的远程控制，实现了两者之间的通信。编写机器人软件控制界面，实时显示履带式机器人的偏转角度。
The tracked mobile robot with special mechanical structure, can adapt to a complex pavementenvironment, and its application is becoming more and more widely. We use the gyroscope toachieve electronic directional and the control of rotation angle based on the robot. Thegyroscope is an internal sensor which can measure the angular velocity of rotating objects. Itis very important that gyroscope does not need the external reference signal. We can usecharacteristic of the gyroscope to achieve more precise navigation and positioning when therobot is in a low environment or we are not familiar with the environment. In the distancemeasurement system of Binocular stereo vision，we use the fish-eye lens and the standard lens.The fish-eye lens axis is perpendicular to the bottom surface of the mobile robot and thestandard lens axis parallel to the direction of mobile robot. So we use the fish-eye lens to findthe target objet easier. At last, we use the standard lens to identify the target objet and achievethe distance measurement with the help of the electronic gyroscope.Because of the uniqueadvantage, it has a very important significance in civil, aerospace, military and others whichrely on the visual information to make decision.
This paper is based on tracked mobile robot. In the distance measurement system ofBinocular stereo vision, we complete the center calibration of the fish-eye image．Accordingto the characteristics of image color, we make the color segmentation twice and get grayscaleimage and binary image．Also we can identify the target object automatically and quicklythrough the image which has been eliminated noise and median filtering．In the experiment,we use centroid of the object to replace of the target object and deduce the formula ofbinocular stereo-vision ranging．At last, we make the analysis and correction of the erroraccording to the experimental data.
As the main part of this article, it designs the ENC-03 gyroscope peripheral circuit inelectronic directional. By using the filtering, amplifier, A/D conversion, it converts the outputanalog signals of the gyroscope into digital signals, and then links with single-chipmicrocomputer and computer. After that, the output voltage value of the gyroscope can beread from the computer. Then the value can be converted into angular velocity and integral, soas to find out the rotation angle of the object. Measuring the Angle of values, verifying theresult, and calculating the maximum error, it meets the accuracy requirement of the ENC-03gyroscope.
Through the wireless network, we can use the computer to control the tracked mobilerobot and complete the communication between the computer and the tracked mobile robot.We write the control software of the tracked mobile robot to display the deflection angle inreal time．
引文
[1]郭宇光.机器人发展的历史、现状、趋势.北京:北京航空航天大学出版社,1989,1-22
[2]宋健.智能控制—超越世纪的目标.国际自动控制联合会(IFAC)第14次代表大会报告,北京,1999.
[3]张毅,罗元,郑太雄.移动机器人技术及其应用.电子工业出版社.2007(9):15-18
[4] John J.Craig.机器人学导论.机械工业出版社.2006(6):22-42
[5]潘进,周易.划时代的超越:机器人技术.北京:北京航空航天大学出版社,2006,1-22
[6]吴小进.履带式移动机器人控制系统设计与实现[D].硕士学位论文,南京理工大学, 2010.
[7]王志文.移动机器人导航技术现状与展望,机器人,2003（9）:470-474
[8] Tanner H G,Kyriakopoulos KJ.Nonholonomic motion planning for mobile manipulators[C].Proceedings of IEEE International Conference on Robotics and Automation , 2000,1233~1238
[9]应科炜,吴文贡,汤达斌,等.ADXRS角速度检测陀螺仪原理及应用.传感器世界.2006（1）:42-45.
[10]张祥德.基于测角的自主移动机器人定位算法,东北大学学报,2002（12）:1143-1145.
[11]周国清，唐晓芳.计算机视觉及应用.计算机用户.1996，13(8):3-6.
[12] A.Bensrhair，M.bertozzi.Stereo Vision-based Feayure Extraction for Vehicle Detection.MachineIntelligence.2001(3):465-470.
[13]陈焱.全方位视觉图像展开和空间定位研究[D].硕士学位论文,浙江工业大学,2006.
[14]王力超,熊超,王晨毅,陆起涌.基于竞争机制的简化双目立体视觉测距算法及系统设计.传感技术学报,2007-1-20.
[15]祝琨，基于双目视觉信息的运动物体实时跟踪.北京：北京交通大学出版社，2008:8-9.
[16] Campion G,Bastin G,D'Andrea-Novol B.Structural properties and classification of kinematic anddynamic models of wheeled mobile robots.IEEE Trans.on Robotics andAutomation,1996,12(1):47-61.
[17] Jain R.Expensive vision.CVGIP image Understanding.1994,5(1):86-88.
[18]石继雨，机器人双目立体视觉技术研究[D].硕士学位论文，哈尔滨工程大学，2003.
[19] Yan Yongjie，Zhu Qidan，Lin Zhuang.Camera Calibration in Binocular StereoVision of MovingRobot.Proceedings of the Sixth World Congress on Intelligent Control and Automation.IEEERobotics and Automation Society Press.2006，34(3):9257-9261.
[20] Yakimovsky Y，Cunningham R.A system for extracting three dimensional measurement from astereo pair of cameras.Computer Graphics and Image Processing.1978, 8(7):195-210.
[21]杨志斌.鱼眼镜头[J].照材荟萃，2008(5):43-44.
[22] Jianhua Wang,Fanhuai Shi,Jing Zhang,A New Calibration Model and Method of Camera LensDistortion[C],Proceedings of the 2006 IEEE/RSJ.Beijing China,International Conference onIntelligent Robots and Systems,October 9–15.2006:5713-5718.
[23] J.Weng.P.Cohen,and M.Herniou,Camera Calibration with Distortion Models and AccuracyEvaluation[J],IEEE Trans.Pattern Analysis and Machine Intelligence,Oct.1992,14(10):965-980.
[24]元伟，全方位视觉移动机器人目标跟踪算法研究[D].硕士学位论文，天津理工大学，2009.
[25]朱锦辉，基于全方位视觉系统的移动机器人自主定位与导航技术研究[D].硕士学位论文，天津理工大学，2008.
[26]刘川辉，视频图像目标检测技术方法研究[D].硕士学位论文，成都理工大学，2010.
[27]吕利娟，视频序列中运动目标的检测与跟踪算法[D].硕士学位论文，西安电子科技大学，2010.
[28]蒲东兵，一种快速车牌定位算法[J].微型电脑应用,2010(10).
[29]黄展鹏,基于分段直方图均衡化技术的图像增强[J].电脑知识与技术，2008(6).
[30]郑小东，基于双目立体视觉的番茄识别与定位技术[J].计算机工程,2004(22).
[31] Ko C C,Chen B M.Development of a Web-based Laboratory for Contol Experiments on aCoupled Tank Apparatus.IEEE Trans.Educ.,2001,44(2):76-86.
[32]胡蕾,几种图像去噪算法的应用分析[J].信息技术,2007(7).
[33] Ma S.A Self-calibration Technique for Active Vision System[J].IEEE Transactions on Roboticsand Automation,1996,12(l):114-120.
[34] Z.Y.Zhang.A flexible new technique for camera calibration[J].IEEE Transactions onpatternAnalysis and Machine Intelligence,2000,22(11):1330-1334
[35] Dan Sheingold,David Krakauer.New iMEMS angular-rate-sensing gyroscope[R].AnalogDialogue,2003:37-03.
[36] Terry Roszhart.The Effects of Collateral Modes on MEMS Gyro Bias,Proc.of SPIEVol.2006:152-164
[37] M.Elwenspoke,R.wiegerink,陶家渠等译.硅微机械传感器[M].北京:中国宇航出版社,2003.
[38]许宜申，单片三轴硅微机械振动陀螺仪研究.高技术通，2006(10).
[39]孙家庆，硅微振动陀螺仪制造误差电补偿方法研究[D].硕士学位论文，南京理工大学，2008.
[40]陈长顺，ADC0832串行模拟转换器及其应用[J].集成电路应用，1992(3).
[41]李鸣华，ATC98C2051在语音合成中的应用[J].计算机与现代，2000(5).
[42]李鹏，轻型飞行模拟器飞行仿真建模研究[D].硕士学位论文，南京航空航天大学，2008.
[43]汪在荣.远程桌面控制服务器研究与实现[D]..硕士学位论文,电子科技大学, 2005.
[44]何牧泓.轻松玩转远程控制[M].重庆:重庆出版社,2002.
[45]高守传,聂云铭,郑静. Visual C++ 6.0开发指南[M].北京:北京出版社,2006.
[46] R.M.Haralick，L.G.ShaPiro.Glossary of Computer Vision Items.Pattem Reeogniton[J]，1991，Vol.24，No.l:69-93.
[47] Chunck Thorp,justin Carlson,Dave Dave Duggins.Safe Robot Driving in Cluttered Environments[J].Proceedings of the 11th International Symposium of Robotics Research.2003,10(1):870－875.
[2]宋健.智能控制—超越世纪的目标.国际自动控制联合会(IFAC)第14次代表大会报告,北京,1999.
[3]张毅,罗元,郑太雄.移动机器人技术及其应用.电子工业出版社.2007(9):15-18
[4] John J.Craig.机器人学导论.机械工业出版社.2006(6):22-42
[5]潘进,周易.划时代的超越:机器人技术.北京:北京航空航天大学出版社,2006,1-22
[6]吴小进.履带式移动机器人控制系统设计与实现[D].硕士学位论文,南京理工大学, 2010.
[7]王志文.移动机器人导航技术现状与展望,机器人,2003（9）:470-474
[8] Tanner H G,Kyriakopoulos KJ.Nonholonomic motion planning for mobile manipulators[C].Proceedings of IEEE International Conference on Robotics and Automation , 2000,1233~1238
[9]应科炜,吴文贡,汤达斌,等.ADXRS角速度检测陀螺仪原理及应用.传感器世界.2006（1）:42-45.
[10]张祥德.基于测角的自主移动机器人定位算法,东北大学学报,2002（12）:1143-1145.
[11]周国清，唐晓芳.计算机视觉及应用.计算机用户.1996，13(8):3-6.
[12] A.Bensrhair，M.bertozzi.Stereo Vision-based Feayure Extraction for Vehicle Detection.MachineIntelligence.2001(3):465-470.
[13]陈焱.全方位视觉图像展开和空间定位研究[D].硕士学位论文,浙江工业大学,2006.
[14]王力超,熊超,王晨毅,陆起涌.基于竞争机制的简化双目立体视觉测距算法及系统设计.传感技术学报,2007-1-20.
[15]祝琨，基于双目视觉信息的运动物体实时跟踪.北京：北京交通大学出版社，2008:8-9.
[16] Campion G,Bastin G,D'Andrea-Novol B.Structural properties and classification of kinematic anddynamic models of wheeled mobile robots.IEEE Trans.on Robotics andAutomation,1996,12(1):47-61.
[17] Jain R.Expensive vision.CVGIP image Understanding.1994,5(1):86-88.
[18]石继雨，机器人双目立体视觉技术研究[D].硕士学位论文，哈尔滨工程大学，2003.
[19] Yan Yongjie，Zhu Qidan，Lin Zhuang.Camera Calibration in Binocular StereoVision of MovingRobot.Proceedings of the Sixth World Congress on Intelligent Control and Automation.IEEERobotics and Automation Society Press.2006，34(3):9257-9261.
[20] Yakimovsky Y，Cunningham R.A system for extracting three dimensional measurement from astereo pair of cameras.Computer Graphics and Image Processing.1978, 8(7):195-210.
[21]杨志斌.鱼眼镜头[J].照材荟萃，2008(5):43-44.
[22] Jianhua Wang,Fanhuai Shi,Jing Zhang,A New Calibration Model and Method of Camera LensDistortion[C],Proceedings of the 2006 IEEE/RSJ.Beijing China,International Conference onIntelligent Robots and Systems,October 9–15.2006:5713-5718.
[23] J.Weng.P.Cohen,and M.Herniou,Camera Calibration with Distortion Models and AccuracyEvaluation[J],IEEE Trans.Pattern Analysis and Machine Intelligence,Oct.1992,14(10):965-980.
[24]元伟，全方位视觉移动机器人目标跟踪算法研究[D].硕士学位论文，天津理工大学，2009.
[25]朱锦辉，基于全方位视觉系统的移动机器人自主定位与导航技术研究[D].硕士学位论文，天津理工大学，2008.
[26]刘川辉，视频图像目标检测技术方法研究[D].硕士学位论文，成都理工大学，2010.
[27]吕利娟，视频序列中运动目标的检测与跟踪算法[D].硕士学位论文，西安电子科技大学，2010.
[28]蒲东兵，一种快速车牌定位算法[J].微型电脑应用,2010(10).
[29]黄展鹏,基于分段直方图均衡化技术的图像增强[J].电脑知识与技术，2008(6).
[30]郑小东，基于双目立体视觉的番茄识别与定位技术[J].计算机工程,2004(22).
[31] Ko C C,Chen B M.Development of a Web-based Laboratory for Contol Experiments on aCoupled Tank Apparatus.IEEE Trans.Educ.,2001,44(2):76-86.
[32]胡蕾,几种图像去噪算法的应用分析[J].信息技术,2007(7).
[33] Ma S.A Self-calibration Technique for Active Vision System[J].IEEE Transactions on Roboticsand Automation,1996,12(l):114-120.
[34] Z.Y.Zhang.A flexible new technique for camera calibration[J].IEEE Transactions onpatternAnalysis and Machine Intelligence,2000,22(11):1330-1334
[35] Dan Sheingold,David Krakauer.New iMEMS angular-rate-sensing gyroscope[R].AnalogDialogue,2003:37-03.
[36] Terry Roszhart.The Effects of Collateral Modes on MEMS Gyro Bias,Proc.of SPIEVol.2006:152-164
[37] M.Elwenspoke,R.wiegerink,陶家渠等译.硅微机械传感器[M].北京:中国宇航出版社,2003.
[38]许宜申，单片三轴硅微机械振动陀螺仪研究.高技术通，2006(10).
[39]孙家庆，硅微振动陀螺仪制造误差电补偿方法研究[D].硕士学位论文，南京理工大学，2008.
[40]陈长顺，ADC0832串行模拟转换器及其应用[J].集成电路应用，1992(3).
[41]李鸣华，ATC98C2051在语音合成中的应用[J].计算机与现代，2000(5).
[42]李鹏，轻型飞行模拟器飞行仿真建模研究[D].硕士学位论文，南京航空航天大学，2008.
[43]汪在荣.远程桌面控制服务器研究与实现[D]..硕士学位论文,电子科技大学, 2005.
[44]何牧泓.轻松玩转远程控制[M].重庆:重庆出版社,2002.
[45]高守传,聂云铭,郑静. Visual C++ 6.0开发指南[M].北京:北京出版社,2006.
[46] R.M.Haralick，L.G.ShaPiro.Glossary of Computer Vision Items.Pattem Reeogniton[J]，1991，Vol.24，No.l:69-93.
[47] Chunck Thorp,justin Carlson,Dave Dave Duggins.Safe Robot Driving in Cluttered Environments[J].Proceedings of the 11th International Symposium of Robotics Research.2003,10(1):870－875.